Dammarane Glycosides from the Root of Machilus yaoshansis

Jul 10, 2012 - New iridoid glycosides from the fruits of Forsythia suspensa and their hepatoprotective activities. Si-Yuan Shao , Ya-Nan Yang , Zi-Min...
0 downloads 0 Views 478KB Size
Article pubs.acs.org/jnp

Dammarane Glycosides from the Root of Machilus yaoshansis Maoluo Gan,†,‡,§ Mingtao Liu,†,§ Lishe Gan,⊥ Sheng Lin,*,† Bo Liu,† Yanling Zhang,† Jiachen Zi,† Weixia Song,† and Jiangong Shi*,† †

State Key Laboratory of Bioactive Substance and Function of Natural Medicines, Institute of Materia Medica, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100050, People’s Republic of China ‡ Institute of Medicinal Biotechnology, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100050, People’s Republic of China ⊥ Institute of Modern Chinese Medicine, College of Pharmaceutical Sciences, Zhejiang University, Hangzhou 310058, People’s Republic of China S Supporting Information *

ABSTRACT: Nine new dammarane triterpene glycosides (1−3 and 8−13) and 12 known analogues have been isolated from an ethanol extract of the roots of Machilus yaoshansis. Compounds 1−7 have an uncommon 20,23-dihydroxydammar-24-en-21-oic acid-21,23-lactone moiety that was previously reported in compounds isolated from Gynostemma pentaphyllum. The configurations of the lactone moieties in 1− 3 were determined by comparison of the experimental ECD spectra of 1−3 and the hydrolysates, 1a and 1b, with the corresponding calculated ECD spectra. On the basis of NMR and ECD data analysis of 1−7, the previously reported C-20 and C-23 configurations of 4−7 and related derivatives from Gynostemma pentaphyllum were revised. In addition, the application of NMR data and Cotton effects to the determination of the relative and absolute configurations of the γ-lactone moiety in 3β,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone derivatives is discussed.

S

G. pentaphyllum were revised. In addition, the application of NMR data and Cotton effects to the determination of the relative and absolute configurations of the γ-lactone moiety in 3β,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone derivatives is discussed.

pecies of the genus Machilus have long been used for the treatment of edema, abdominal distension, pain, and inflammation in China and Southeast Asia.1 As part of a program to assess the chemical and biological diversity of Machilus species,2 we focused our study on Machilus yaoshansis S. Lee et F. N. Wei, a plant that is widely distributed in the south of China and used as a folk medicine by the ethnic Zhuang in Guangxi Province for the treatment of rheumatism. Our previous studies on the bark3 and the root4 of M. yaoshansis led to the isolation and characterization of several unusual cucurbitane derivatives and spirolactones with cytotoxic activities. Continuing examination of the root extract has resulted in the characterization of nine new (1−3 and 8−13) and 12 known dammarane glycosides. Herein, we discuss the detailed structural determination of the isolates by extensive spectroscopic analysis, including 1D and 2D NMR and electronic circular dichroism (ECD). Compounds 1−7 possess an unusual 20,23-dihydroxydammar-24-en-21-oic acid-21,23lactone moiety that was previously reported in compounds isolated from Gynostemma pentaphyllum. The C-20 and C-23 configurations of the 20,23-dihydroxydammar-24-en-21-oic acid-21,23-lactone moieties of 1−3 were established by comparison of the ECD spectra calculated using the timedependent density functional theory (TDDFT) with the experimental ECD spectra. On the basis of detailed NMR and ECD data analysis of 1−7, the previously reported C-20 and C-23 configurations of 4−7 and related derivatives from © 2012 American Chemical Society and American Society of Pharmacognosy



RESULTS AND DISCUSSION Compound 1 was obtained as an amorphous powder, and its molecular formula was determined to be C46H74O16 by positive-ion HRESIMS data at m/z 905.4832 [M + Na]+ (calcd for C46H74O16Na, 905.4874), combined with the NMR data (Tables 1 and 3). The IR spectrum suggested the presence of OH (3383 cm−1) and γ-lactone (1760 cm−1) functionalities. The 1H NMR spectrum showed resonances assignable to seven tertiary methyl groups between δH 0.76 and 1.69 and five deshielded methines (anomeric, olefinic, and oxygenated) at δH 6.13 (brs, H-1″), 5.59 (d, J = 8.8 Hz, H-24), 5.47 (ddd, J = 8.8, 7.6, 6.0 Hz, H-23), 4.92 (d, J = 5.2 Hz, H-1′), and 5.02 (d, J = 7.2 Hz, H-1″′). The 1H NMR spectrum also showed partially overlapped signals due to oxymethine and oxymethylene protons between δH 3.65 and 4.73. 13C NMR and DEPT spectra showed 46 carbon resonances, of which three were attributed to anomeric carbons [δC 105.2 (C-1″′), 104.8 (C-1′), and 102.1 (C-1″)], two were attributed to a trisubstituted Received: May 1, 2012 Published: July 10, 2012 1373

dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products

Article

3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]α-L-arabinopyranoside. The configuration of the tetracyclic nucleus in 1 was proposed to be identical to that in the natural dammarane derivatives.6 The chemical shift and splitting pattern of H-3 (δH 3.29, dd, J = 12.0 and 4.0 Hz) in the 1H NMR spectrum indicated that it was axially α-oriented.6a The configuration of the γ-lactone in the side chain was established by analyses of the NOESY and ECD data of 1 and its hydrolysates (1a and 1b), in combination with the theoretical ECD spectra based on TDDFT, a powerful tool for the configuration assignment of natural products.7 In the NOESY spectrum, correlations of H-23/H-17 and H-22a revealed that these protons were cofacial on the γ-lactone ring, while the correlation of H-24/H-22b indicated that these protons were cofacial on the opposite side of the ring. Acid hydrolysis of 1 with 1 M HCl at 60 °C produced two isomers, 1a and 1b. The NOESY spectrum of 1a showed correlations of H-23/H-17 and H-22a and of H-22b/H-24 and OH-20. In contrast, 1b did not give the corresponding correlations in its NOESY spectrum (Supporting Information, Figures S20 and S26). This indicated that 1a had the same configuration as 1, but 1b was an epimer of 1a with the opposite configuration at C-20 or C-23. This was supported by the similarity of the Cotton effects in the ECD spectra of 1 and 1a, while the ECD spectrum of 1b displayed Cotton effects opposite those of 1 and 1a (Figure 1). The ECD spectra of 1a and the (20R,23S)-isomer were calculated using TDDFT at the B3LYP/6-311++G(2d,2p) level. The calculated ECD spectrum of 1a matched the experimental spectra of 1a and 1. This indicated the (20S,23R) configuration for 1 and 1a. Therefore, compound 1 was determined to be (3β,20S,23R)-3,20,23-trihydroxydammar-24en-21-oic acid-21,23-lactone 3-O-[α-L-rhamnopyranosyl-(1→2)][β-D-xylopyranosyl-(1→3)]-α-L-arabinopyranoside. Compound 2 is an isomer of 1, as indicated by spectroscopic data. Comparison of the NMR data of 2 and 1 indicated that H22a, H-22b, H-23, and H-24 in 2 were shifted by ΔδH −0.10, −0.18, +0.24, and −0.18, respectively, while C-20, C-22, C-23, and C-24 were shifted by ΔδC +2.1, −1.8, +1.1, and −1.4, respectively. This suggested that 2 was the C-23 epimer of 1, as supported by the ECD spectrum of 2, which displayed Cotton effects opposite those of 1. Acid hydrolysis of 2 generated the same compounds as those from 1, including the same sugars and 1a and 1b. ECD calculations of 1b and the (20R,23R)isomer demonstrated that the calculated ECD curve for 1b was consistent with the experimental ECD spectra of 1b and 2. This confirmed that 2 was the C-23 epimer of 1. Therefore, compound 2 was determined to be (3β,20S,23S)-3,20,23-trihydroxydammar24-en-21-oic acid-21,23-lactone 3-O-[α-L-rhamnopyranosyl(1→2)]-[β-D-xylopyranosyl(1→3)]-α-L-arabinopyranoside. Compound 3 exhibited IR and NMR spectroscopic features similar to those of 2. Comparison of the NMR, HRESIMS, and ECD data of 3 and 2 (Tables 1 and 3 and Experimental Section) indicated that the α-L-arabinopyranosyl and α-Lrhamnopyranosyl moieties in 2 were replaced by a β-Dglucopyranosyl moiety in 3. This was confirmed by acid hydrolysis of 3 and subsequent GC analysis of the hydrolysate, according to the same protocol as that described for 1. In the HMBC spectrum of 3, correlations of H-1′/C-3 and H-1″/C-3′ revealed the linkage of the two sugar units. Thus, compound 3 was assigned as (3β,20S,23S)-3,20,23-trihydroxydammar-24-en21-oic acid-21,23-lactone 3-O-β-D-xylopyranosyl-(1→3)-β-Dglucopyranoside.

double bond (δC 138.5 and 125.4), and one was attributed to a lactone carbonyl carbon (δC 179.4). These spectroscopic data suggested that 1 was a triglycosidic triterpenoid with γ-lactone and olefinic subunits. The sugars obtained by acid hydrolysis of 1 were identified as L-arabinopyranose, D-xylopyranose, and L-rhamnopyranose by GC analysis of the trimethylsilyl-L-cysteine derivatives of the hydrolysate of 1 and the authentic sugars.5 The structure of 1 was finalized by analysis of the 2D NMR data. The HSQC experiment allowed for the assignments of the proton and protonated carbon resonances in the NMR spectra of 1. In the 1H−1H COSY spectrum, the cross-peaks of H2-1/ H2-2/H-3; H-5/H2-6/H2-7; H-9/H2-11/H2-12/H-13/H-17/ H2-16/H2-15; and H2-22/H-23/H-24 demonstrated the presence of vicinal coupling systems. Coupling constants of the anomeric protons indicated α configurations for the arabinopyranosyl (J1′,2′ = 5.2 Hz) and rhamnopyranosyl (J1″,2″ ≈ 0 Hz) units and a β configuration for the xylopyranosyl (J1″′,2″′ = 7.2 Hz) unit.6 In the HMBC spectrum, two- and threebond correlations of H3-18/C-7, C-8, C-9, and C-14; H3-19/C-1, C-5, C-9, and C-10; H3-26, H3-27/C-24 and C-25; H3-28 and H3-29/C-3, C-4, and C-5; H3-30/C-8, C-13, C-14, and C-15; and H2-22/C-17, C-20, C-21, C-23, and C-24, together with the chemical shifts of these protons and carbons and the molecular formula, revealed that 1 had a 3,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone nucleus.6a,b In addition, HMBC correlations of H-1′/C-3, H-1″/C-2′, and H-1″′/ C-3′ demonstrated that the α-L-arabinopyranosyloxy unit was located at C-3 and that the α-L-rhamnopyranosyloxy and β-D-xylopyranosyloxy units were linked to C-2′ and C-3′ of the arabinopyranosyloxy unit, respectively. Thus, 1 was determined to be 3,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone 1374

dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products

Article

Table 1. 1H NMR Data for Compounds 1−3, 8, and 9 in Pyridine-d5 (δ, mult., J in Hz)a no.

1

2

3

1 2 3 5 6 7 9 11 12 13 15 16 17 18 19 21

1.53 2.05 3.29 0.73 1.46 1.50 1.30 1.46 2.40 2.07 1.61 2.02 2.52 0.91 0.76

m, 0.81 m m, 1.86 qd (13.2, 4.0) dd (12.0, 4.0) d (12.0) m, 1.36 m m, 1.21 m m m, 1.22 m m, 1.34 m m m, 1.12 m m, 1.69 m td (10.0, 4.8) s s

1.56 2.08 3.31 0.74 1.48 1.50 1.31 1.50 2.51 1.85 1.60 2.05 2.71 0.99 0.82

m, 0.78 m m, 1.80 m dd (12.0, 4.0) d (11.2) m, 1.38 m m, 1.23 m m m, 1.30 m brd (12.0), 1.47 m m m, 1.10 m m, 1.30 m td (10.4, 5.6) s s

1.48 2.21 3.37 0.72 1.50 1.52 1.28 1.50 2.51 1.84 1.58 2.07 2.71 0.98 0.79

m, 0.76 m m, 1.82 m dd (12.0, 4.0) d (12.0) m, 1.37 m m, 1.22 m m m, 1.29 m brd (12.0), 1.46 m m m, 1.09 m m, 1.31 m td (10.5, 4.5) s s

22

2.70 2.30 5.47 5.59 1.62 1.69 1.19 1.11 0.90 4.92 4.67 4.29 4.48 4.33 3.81

dd (13.6, 7.2) dd (13.6, 6.0) ddd (8.8, 7.6, 6.0) d (8.8) s s s s s d (5.2) dd (7.2, 5.2) dd (7.2, 3.2) ddd (4.8, 3.2, 2.4) dd (11.2, 4.8) dd (11.2, 2.4)

2.60 dd (13.2, 5.6) 2.12 dd (13.2, 10.0) 5.71 ddd (10.0, 8.8, 5.6) 5.41d (8.8) 1.67 s 1.63 s 1.20 s 1.14 s 0.91 s 4.92 d (5.2) 4.67 dd (7.2, 5.2) 4.31 dd (7.2, 3.2) 4.48 ddd (4.8, 3.2, 2.4) 4.33 dd (11.2, 4.8) 3.82 dd (11.2, 2.4)

2.59 2.10 5.70 5.40 1.67 1.63 1.30 1.00 0.90 4.91 4.06 4.22 4.11 3.95

dd (13.0, 5.5) dd (13.0, 9.5) ddd (9.5, 9.0, 5.5) d (9.0) s s s s s d (8.0) dd (8.5, 8.0) dd (8.5, 8.5) dd (8.5, 8.5) ddd (8.5, 5.5, 2.0)

4.53 4.33 5.26 4.01 4.14 4.14 4.29 3.70

dd (12.5, 2.0) dd (12.5, 5.5) d (8.0) dd (8.0, 8.0) dd (8.0, 8.0) ddd (10.0, 8.0, 5.0) dd (11.0, 5.0) dd (11.0, 10.0)

23 24 26 27 28 29 30 1′ 2′ 3′ 4′ 5′a 5′b 6′a 6′b 1″ 2″ 3″ 4″ 5″a 5″b 6″ 1″′ 2″′ 3″′ 4″′ 5″′a 5″′b

6.13 4.73 4.57 4.27

brs dd (3.6, 1.6) dd (9.6, 3.6) dd (9.6, 9.6)

6.14 4.73 4.57 4.28

brs dd (3.2, 1.2) dd (9.6, 3.2) dd (9.6, 9.6)

4.59 1.62 5.02 3.92 4.09 4.11 4.30 3.65

dq (9.6, 6.0) d (6.0) d (7.2) dd (8.0, 7.2) dd (8.4, 8.0) ddd (10.8, 8.4, 4.8) dd (10.8, 4.8) dd (10.8, 10.8)

4.59 1.62 5.02 3.93 4.08 4.11 4.30 3.66

m d (6.0) d (7.6) dd (8.0, 7.6) dd (8.4, 8.0) ddd (10.8, 8.4, 4.8) dd (10.8, 4.8) dd (10.8, 10.8)

9b

8 1.52 m, 0.82 m 2.04 m, 1.85 m 3.28 dd (11.5, 4.0) 0.74 brd (12.0) 1.45 m, 1.36 m 1.52 m, 1.23 m 1.33 m 1.49 m, 1.20 m 2.22 brd (11.0), 1.41 m 2.12 m 1.65 m, 1.12 m 2.00, 1.94 m 2.30 m 0.95 s 0.77 s 4.06 d (11.0) 4.00 d (11.0) 2.08 m 1.95 m 2.49 m, 2.40 m 5.33 t (6.0) 1.66 s 1.63 s 1.19 s 1.11 s 0.99 s 4.92 d (5.0) 4.66 dd (7.0, 5.0) 4.30 dd (7.0, 3.0) 4.48 ddd (4.5, 3.0, 2.5) 4.32 dd (12.0, 4.5) 3.83 d (12.0, 2.5)

1.51 2.04 3.26 0.72 1.44 1.51 1.29 1.37 2.13 2.05 1.61 1.92 2.21 0.92 0.75 4.38 4.01 2.09 1.90 2.43 5.27 1.64 1.62 1.17 1.10 0.93 4.91 4.65 4.30 4.47 4.33 3.83

m, 0.80 m m, 1.83 m dd (11.5, 4.0) brd (12.0) m, 1.35 m m, 1.21 m m m, 1.11 m m, 1.30 m m m, 1.08 m m, 1.88 m m s s d (11.0) d (11.0) m m m, 2.31 m t (7.0) s s s s s d (5.5) dd (7.0, 5.5) dd (7.0, 3.0) ddd (5.0, 3.0, 2.5) dd (12.0, 5.0) dd (12.0, 2.5)

6.13 4.73 4.57 4.27

brs brs dd (9.0, 3.5) dd (9.0, 9.0)

6.11 4.72 4.57 4.26

brs brs dd (9.0, 3.5) dd (9.0, 8.5)

4.59 1.62 5.01 3.93 4.09 4.11 4.30 3.65

m d (6.0) d (7.5) dd (8.0, 7.5) dd (8.0, 8.0) ddd (9.5, 8.0, 4.5) dd (11.0, 4.5) dd (11.0, 9.5)

4.59 1.61 5.00 3.92 4.08 4.11 4.30 3.66

m d (6.0) d (7.0) dd (8.0, 7.0) dd (8.0, 8.0) ddd (9.5, 8.0, 5.0) dd (11.0, 5.0) dd (11.0, 9.5)

a1

H NMR data were measured at 400 MHz for 1 and 2 and 500 MHz for 3, 8, and 9. The assignments were based on DEPT, 1H−1H COSY, HSQC, and HMBC experiments. bData for Glc-21: δ 5.04 d (7.5 Hz, H-1″″), 4.08 dd (8.5 and 7.5 Hz, H-2″″), 4.20 t (8.5 Hz, H-3″″), 4.22 t (8.5 Hz, H-4″″), 3.97 m (H-5″″), 4.55 dd (11.5 and 2.5 Hz, H-6″″a), and 4.37 dd (11.5 and 4.0 Hz, H-6″″b).

3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→ 3)]-β-D-6-O-acetylglucopyranoside, and (3β,20R,23R)-3,20,23trihydroxydammar-24-en-21-oic acid-21,23-lactone 3-O-[α-Lrhamnopyranosyl-(1→2)]-[β- D -xylopyranosyl-(1→3)]-β- Dglucopyranoside, respectively, which were isolated from G. pentaphyllum. 6b These compounds, along with five other derivatives6a,b,8 from G. pentaphyllum, represent glycosides with the 3β,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone aglycone moiety. However, the initial assignment of

The IR, NMR, and HRESIMS spectroscopic data for 4, 5, 6, and 7 (Supporting Information, Tables S2 and S3) were identical to those of (3β,20S,23S)-3,20,23-trihydroxydammar24-en-21-oic acid-21,23-lactone 3-O-[α-L-rhamnopyranosyl(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-6-O-acetylglucopyranoside, (3β,20S,23S)-3,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β- D -glucopyranoside, (3β,20R,23R)3,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone 1375

dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products

Article

Table 2. 1H NMR Data for Compounds 10−13 in Pyridine-d5 (δ, mult., J in Hz)a no. 1 2 3 5 6 7 9 11 12 13 15 16 17 18 19 21 22 23 24 26 27 28 29 30 1′ 2′ 3′

10

11b

12c

2.63 m, 0.71 m 2.16 m, 1.67 m 3.38 dd (11.6, 3.6) 1.18 m 1.85 m, 1.65 m 1.62 m, 1.35 m 1.70 m 1.72 m, 1.17 m 2.17 m, 1.37 m 2.05 m 1.59 m, 1.17 m 2.02 m, 1.92 m 2.27 m 0.86 s 10.03 s 4.01 d (10.4) 3.94 d (10.4) 2.05 m, 1.93 m 2.45 m, 2.38 m 5.32 t (7.2) 1.67 s 1.61 s 1.33 s 0.91 s 0.98 s 4.75 d (7.6) 4.54 dd (8.8, 7.6)

2.45 m, 0.63 m 2.23 m, 1.62 m 3.38 dd (12.0, 4.0) 1.08 brs 1.85 m, 1.60 m 1.58 m, 1.31 m 1.65 m 1.58 m, 1.06 m 2.12 m, 1.29 m 1.97 m 1.56 m, 1.12 m 2.03 m, 1.92 m 2.23 m 0.94 s 10.25 s 4.34 d (11.0) 3.97 d (11.0) 2.04 m, 1.86 m 2.44 m, 2.31 m 5.27 t (7.0) 1.62 s 1.61 s 1.27 s 1.11 s 0.83 s 4.86 d (7.5) 4.22 d (8.0, 7.5)

1.60 m, 0.96 m 2.21 m, 1.87 m 3.32 dd (12.0, 4.0) 0.76 brd (11.0) 1.47 m, 1.38 m 1.52 m, 1.23 m 1.38 m 1.49 m, 1.20 m 2.24 m, 1.42 m 2.13 m 1.66 m, 1.13 m 2.00 m, 1.94 m 2.30 m 0.96 s 0.79 s 4.06 d (11.0) 3.99 d (11.0) 2.11 m, 1.96 m 2.49 m, 2.39 m 5.33 t (7.0) 1.66 s 1.62 s 1.24 s 1.15 s 0.98 s 4.84 d (7.5) 4.21 dd (8.0, 7.5)

4.18 dd (8.8, 3.6) 4.16 dd (8.0, 8.0)

13d

no.

1.38 m, 0.75 m 2.23 m, 1.81 m 3.34 dd (11.5, 4.0) 0.67 brd (11.5) 1.45 m, 1.35 m 1.47 m, 1.19 m 1.24 m 1.35 m, 1.14 m 2.26 m, 1.35 m 2.03 m 1.61 m, 1.08 m 1.92 m, 1.81 m 2.31 m 0.92 s 0.73 s 4.29 d (11.0) 4.08 d (11.0) 2.01 m, 1.82 m 2.49 m, 2.30 m 5.27 t (7.0) 1.63 s 1.65 s 1.22 s 1.15 s 0.92 s 4.88 d (7.5) 4.23 dd (8.0, 7.5) 4.15 dd (9.0, 8.0) 4.17 dd (8.5, 8.0)

10

11b

12c

13d

4′

4.43 brs

3.98 dd (8.0, 8.0)

3.82 dd (9.0, 9.0) 4.00 dd (8.5, 8.5)

5′a

4.25 dd (11.6, 4.0)

3.89 ddd (80, 5.5, 2.0)

3.97 ddd (9.0, 5.5, 2.0)

3.90 ddd (8.5, 5.5, 2.0)

6′a

4.49 dd (12.5, 2.0)

4.82 dd (12.0, 2.0)

4.50 dd (12.0, 2.0)

6′b

4.25 dd (12.6, 5.5)

4.70 dd (12.0, 5.5)

4.26 dd (12.0, 5.5)

5′b 3.80 brd (11.6)

1″

5.26 d (7.6)

6.44 brs

6.44 brs

2″

4.00 dd (8.0, 7.6) 4.79 dd (3.5, 1.5)

6.41 brs

4.78 brd (3.5)

4.80 brd (3.5)

3″

4.14 dd (8.4, 8.0) 4.57 dd (9.5, 3.5)

4.58 dd (9.5, 3.5) 4.59 dd (9.0, 3.5)

4″

4.19 dd (10.0, 8.0, 4.8)

4.26 dd (9.5, 9.5)

4.29 dd (9.5, 9.5) 4.29 dd (9.0, 9.0)

5″a 4.34 dd (11.2, 4.8)

4.69 dq (9.5, 6.0)

4.74 dq (9.5, 6.0) 4.75 dq (9.0, 6.0)

6″ 1″′

1.64 d (6.0) 4.99 d (8.0)

1.67 d (6.0) 4.98 d (7.5)

2″′

3.95 dd (8.0, 8.0)

3.96 dd (8.5, 7.5) 3.96 dd (8.5, 8.0)

3″′

4.07 dd (8.0, 8.0)

4.07 dd (9.0, 8.5) 4.07 dd (8.5, 8.5)

4″′

4.10 ddd (10.0, 8.0, 5.0)

4.11 ddd (10.5, 9.0, 5.0)

4.10 m

5″′a

4.27 dd (11.0, 5.0)

4.27 dd (11.0, 5.0)

4.27 m

5″′b

3.68 dd (11.0, 10.0)

3.68 dd (11.0, 10.5)

3.68 dd (11.0, 10.0)

5″b 3.70 dd (11.2, 10.0) 1.67 d (6.0) 4.99 d (8.0)

a1 H NMR data were measured at 400 MHz for 10 and 500 MHz for 11−13. The assignments were based on DEPT, 1H−1H COSY, HSQC, and HMBC experiments. bData for Glc-21 of 11: δ 5.02 d (7.5 Hz, H-1″″), 4.08 dd (8.0 and 7.5 Hz, H-2″″), 4.21 dd (8.0 and 9.0 Hz, H-3″″), 4.23 t (9.0 Hz, H-4″″), 3.97 m (H-5″″), 4.55 dd (12.0 and 2.0 Hz, H-6″″a), and 4.36 dd (12.0 and 5.5 Hz, H-6″″b). cData for OAc of 12: δ 2.04 s. dData for inner Glc-21 of 13: δ 4.95 d (8.0 Hz, H-1″″), 4.04 t (8.0 Hz, H-2″″), 4.17 t (8.0 Hz, H-3″″), 4.08 dd (8.0 and 9.0 Hz, H-4″″), 4.10 m (H-5″″), 4.87 brd (12.0 Hz, H-6″″a), and 4.22 m (H-6″″b); for terminal Glc-21 of 13: δ 5.00 d (8.0 Hz, H-1″″′),4.04 t (8.0 Hz, H-2″″′), 4.22 dd (8.0 and 9.0 Hz, H-3″″′), 4.20 dd (8.0 and 9.0 Hz, H-4″″′), 3.90 m (H-5″″′), 4.49 dd (12.0 and 2.0 Hz, H-6″″′a), and 4.35 dd (12.0 and 5.0 Hz, H-6″″′b).

products (1a and 1b) as those from 1−3. These results also indicated that the configuration of the reported (3β,20S,23S)and (3β,20R,23R)-3,20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone derivatives6b,8 should be revised as (3β,20S,23R) and (3β,20S,23S), respectively. This was further supported by the [α]20D values of 4−7, 1a, and 1b, which were consistent with those of the reported compounds having the same gross structures.6b,10 In addition, the configurations at C-20 of the (20R,23ξ)- and (20S,23ξ)-derivatives6a,10 did not match the presented structures, and on basis of the above revision, the configurations had to be revised as (20S,23R) and (20S,23S), respectively. Therefore, 4, 5, 6, and 7 were defined as (3β,20S,23R)-3,20,23-trihydroxydammar-24-en-21-oic acid21,23-lactone 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-6-O-acetylglucopyranoside, (3β,20S,23R)-3, 20,23-trihydroxydammar-24-en-21-oic acid-21,23-lactone 3-O[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-βD-glucopyranoside, (3β,20S,23S)-3,20,23-trihydroxydammar24-en-21-oic acid-21,23-lactone 3-O-[α-L-rhamnopyranosyl(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-6-O-acetylglucopyranoside, and (3β,20S,23S)-3,20,23-trihydroxydammar-24-en21-oic acid-21,23-lactone 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-Dxylopyranosyl-(1→3)]-β-D-glucopyranoside, respectively. The sugar

(20S) and (20R) configurations for two reported C-20 epimers,6a made by comparison of the chemical shifts of C-13 and C-17 in these compounds with those in known ginseng saponins,9 was ambiguous. This is because the substitution patterns of these epimers differed significantly from those of the reference ginseng saponins, and the configurations of the other reported compounds were determined by comparison of their NMR data with those of the two C-20 epimers. This, together with the configuration assignments of 1−3, prompted us to reexamine the configurations of the reported compounds by comparing the NMR data of 1−3 with those of 4−7 and the reported compounds.6a,b,8 Without exception, the data for the aglycone moiety in 1 were consistent with those of the aglycone moieties in 4 and 5 and the (20S,23S)-aglycone moiety in the reported compounds. In addition, the data for the aglycone moieties in 2 and 3 were similar to those of the aglycone moieties in 6 and 7 and the (20R,23R)-aglycone moiety in the reported glycosides. This, in combination with the configuration assignments of 1−3 made on the basis of NOESY and ECD data, indicated that 4 and 5 had the same aglycone as 1 and that 6 and 7 had the same aglycone as 2 and 3. This was proved by the Cotton effects in the ECD spectra of 4−7, as well as by acid hydrolysis of 4−7, which generated the same 1376

dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products

Article

Table 3. 13C NMR Data for Compounds 1−3 and 8−13 in Pyridine-d5a no.

1

2

3

8

9b

10

11c

12d

13e

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 1′ 2′ 3′ 4′ 5′ 6′ 1″ 2″ 3″ 4″ 5″ 6″ 1″′ 2″′ 3″′ 4″′ 5″′

39.6 26.8 88.3 39.8 56.6 18.4 35.7 40.8 51.2 37.1 21.8 27.3 43.3 50.7 31.7 25.8 45.8 15.6 16.5 79.0 179.4 40.8 74.1 125.4 138.5 25.6 18.1 27.9 16.8 16.6 104.8 74.7 81.5 68.3 64.8

39.8 26.8 88.2 39.6 56.6 18.4 35.7 40.7 51.2 37.1 21.8 28.0 45.0 50.2 31.7 26.2 45.3 15.6 16.6 81.1 178.3 39.0 75.2 124.0 139.4 25.6 18.1 27.9 16.8 16.2 104.7 74.7 81.5 68.3 64.8

39.8 26.8 88.3 39.6 56.6 18.5 35.7 40.8 51.2 37.0 21.9 28.1 41.8 50.5 31.8 24.8 46.3 15.8 16.6 76.6 66.8 36.6 23.3 126.2 130.8 25.8 17.7 28.0 16.7 16.8 104.8 74.7 81.6 68.3 64.8

39.8 26.8 88.3 39.5 56.5 18.4 35.8 40.7 51.1 37.0 21.8 27.8 41.7 50.5 31.6 24.7 46.1 15.8 16.6 76.4 76.3 36.6 23.3 126.0 130.8 25.8 17.8 27.9 16.6 16.8 104.7 74.6 81.4 68.2 64.8

33.5 27.7 87.5 40.1 54.6 17.7 34.7 40.4 52.9 52.8 22.4 27.9 41.6 50.3 32.1 24.7 46.1 16.6 205.7 76.4 66.6 36.6 23.2 126.2 130.2 25.8 17.7 26.5 15.9 17.2 107.4 71.8 83.5 69.4 67.1

102.1 72.5 72.6 74.0 70.1 18.6 105.2 74.6 77.7 70.7 67.0

102.1 72.5 72.6 73.9 70.1 18.6 105.1 74.6 77.7 70.9 67.0

39.7 26.7 89.1 39.3 56.4 18.4 35.7 40.7 51.1 37.0 21.8 27.9 45.0 50.2 31.7 26.2 45.3 15.6 16.5 81.1 178.3 39.0 75.3 124.0 139.4 25.6 18.1 28.0 16.8 16.2 106.5 74.7 87.8 69.6 78.1 62.7 106.3 75.2 78.2 70.9 67.4

102.1 72.5 72.6 74.0 70.1 18.6 105.1 74.6 77.7 70.9 67.0

102.0 72.4 72.5 73.9 70.6 18.6 105.0 74.5 77.6 70.9 67.0

106.9 75.3 78.2 71.1 67.2

33.6 27.7 88.1 40.4 54.9 17.6 34.6 40.0 52.8 52.7 22.3 27.9 41.6 50.3 32.0 24.7 46.0 16.0 205.5 76.3 76.2 36.5 23.3 126.0 130.9 25.8 17.8 26.3 16.5 17.2 104.9 76.9 87.6 69.9 78.2 62.6 101.8 72.4 72.6 73.9 69.7 18.6 104.9 74.8 78.3 70.6 67.3

39.7 26.8 88.6 39.7 56.8 18.5 35.7 40.8 51.2 37.1 21.9 28.1 41.8 50.5 31.7 24.8 46.3 15.8 16.6 76.6 66.8 36.6 23.3 126.2 130.8 25.8 17.7 27.8 16.7 16.8 105.0 76.6 87.8 69.8 74.4 64.1 101.8 72.4 72.5 73.9 69.9 18.6 104.9 74.8 78.3 70.6 67.3

39.7 26.9 88.9 39.7 56.6 18.5 35.6 40.7 51.0 37.0 21.8 27.7 41.9 50.3 31.5 24.9 46.5 15.8 16.6 76.3 77.8 35.8 23.1 126.0 130.9 25.8 17.8 27.9 16.7 16.9 104.9 77.0 88.2 69.8 78.0 62.6 101.8 72.4 72.5 73.9 69.7 18.6 105.0 74.8 78.2 70.6 67.3

a

Data were measured at 100 MHz for 1, 2, and 10 and 125 MHz for 3, 8, 9, and 11−13. The assignments were based on DEPT, 1H−1H COSY, HSQC, and HMBC experiments. bData for Glc-21 of 9: δ 106.2 (C-1″″), 75.5 (C-2″″), 78.7 (C-3″″), 71.1 (C-4″″), 78.6 (C-5″″), 62.8 (C-6″″). cData for Glc-21of 11: δ 106.2 (C-1″″), 75.5 (C-2″″), 78.7 (C-3″″), 71.1 (C-4″″), 78.6 (C-5″″), 62.8 (C-6″″). dData for OAc of 12: δ 170.6, 20.8. eData for inner Glc-21of 13: δ 106.3 (C-1″″), 75.4 (C-2″″), 78.5 (C-3″″), 71.7 (C-4″″), 77.0 (C-5″″), 70.1 (C-6″″); for terminal Glc-21 of 13: δ 105.2 (C-1″″′), 75.2 (C-2″″′), 78.3 (C-3″″′), 71.5 (C-4″″′), 78.3 (C-5″″′), 62.6 (C-6″″′).

the γ-lactone moiety in the 3β,20,23-trihydroxydammar-24-en21-oic acid-21,23-lactone derivatives. In compounds with OH-20 and H-23 in the trans orientation [(20S,23R)- or (20R,23S)isomers], the H-24 resonance was deshielded by OH-20. Thus, the chemical shift of H-24 (δH 5.60 ± 0.02 for the glycosides in pyridine-d5; δH 5.31 and 5.26 ± 0.01 for the aglycone in acetone-d6 and CDCl3, respectively) was larger than that of

units in 4−7 were confirmed by acid hydrolysis and subsequent GC analysis of the hydrolysates using the protocol described earlier, and their linkages were confirmed by 2D NMR data. Detailed NMR data analysis of 1−7 and the reported compounds6a,b,8,10 indicated that the readily distinguishable H-23 and H-24 resonances in the 1H NMR spectra should be applicable to the determination of the relative configuration of 1377

dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products

Article

Figure 1. (a) Experimental ECD spectra of 1 and 1a and the calculated ECD spectra of 1a and the (20R,23S)-isomer of 1a. (b) Experimental ECD spectra of 2 and 1b and the calculated ECD spectra of 1b and the (20R,23R)-isomer of 1b.

H-23 (δH 5.47 ± 0.02 for the glycosides in pyridine-d5; δH 5.23 and 5.12 for the aglycone in acetone-d6 and CDCl3, respectively). On the other hand, in the compounds with OH-20 and H-23 in the cis orientation [(20S,23S)- or (20R,23R)-isomers], the H-23 resonance was deshielded by OH-20 and the chemical shift of H-23 (δH 5.70 ± 0.02 for the glycosides in pyridine-d5; δH 5.31 and 5.34 for the aglycone in acetone-d6 and CDCl3, respectively) was larger than that of H-24 (δH 5.42 ± 0.02 for the glycosides in pyridine-d5; δH 5.24 and 5.19 for the aglycone in acetone-d6 and CDCl3, respectively). To eliminate errors, the chemical shift difference between the H-24 and H-23 resonances (ΔδH = δH‑24 − δH‑23) is proposed to be used in determination of the relative configuration of the γ lactone moiety in the 3β,20,23trihydroxydammar-24-en-21-oic acid 21,23-lactone derivatives. For compounds with OH-20 and H-23 in the trans orientation, the ΔδH value is positive (approximately +0.13 ± 0.02 ppm for the glycosides in pyridine-d5; +0.08 ± 0.01 and +0.14 ± 0.01 ppm for the aglycone in acetone-d6 and CDCl3, respectively). Conversely, for the compounds with OH-20 and H-23 in the cis orientation, the ΔδH value is negative (around −0.30 ± 0.02 ppm for the glycosides in pyridine-d5; −0.07 ± 0.01 and −0.15 ± 0.01 ppm for the aglycone in acetone-d6 and CDCl3, respectively). In addition, the compounds with OH-20 and H-23 in the trans and cis orientations showed different chemical shifts for the C-20, C-21, C-22, C-23, C-24, and C-25 resonances in the 13C NMR spectra. In pyridine-d5, for the glycosides with OH-20 and H-23 in the trans orientation, C-20, C-23, and C-25 were shielded by ΔδC −2.1 ± 0.1, −1.1 ± 0.1, and −1.0 ± 0.1 ppm, respectively, as compared with those for the compounds with OH-20 and H-23 in the cis orientation, whereas C-21, C-22, and C-24 were deshielded by ΔδC +1.1 ± 0.2, +1.6 ± 0.2, and +1.4 ± 0.2 ppm. To eliminate errors, the chemical shift difference between the readily distinguishable C-25 and C-24 resonances (ΔδC = δC‑25 − δC‑24) is suggested to be used in assignment of the relative configuration of the γ-lactone moiety in the 3β,20,23-trihydroxydammar-24-en-21oic acid-21,23-lactone derivatives. The ΔδC value is smaller than +14.0 ppm for the glycosides with OH-20 and H-23 in the trans orientation (+13.1 ± 0.1 ppm, pyridine-d5), but for the glycosides with OH-20 and H-23 in the cis orientation, the ΔδC value is larger than +14.0 ppm (+15.6 ± 0.2 ppm, pyridine-d5). Similar chemical shift differences are observed for the aglycones 1a and 1b in CDCl3 (ΔδC: +17.0 ± 0.2 ppm for 1a and +18.1 ± 0.1 ppm for 1b) (Supporting Information, Table S1).10

Furthermore, comparison of the Cotton effects in the experimental ECD spectra of 1−7 with those in the calculated ECD spectra of 1a and the (20R,23S)-isomer and 1b and the (20R,23R)-isomer revealed that the absolute configurations of the γ-lactone moiety in the 3β,20,23-trihydroxydammar-24-en21-oic acid 21,23-lactone derivatives could be determined by the Cotton effects. For the compounds with OH-20 and H-23 in the trans orientation, the (20S,23R)-isomers showed positive Cotton effects at 227 ± 1 nm and negative effects at 206 ± 2 nm (calculated at 231 and 208 nm), while the (20R,23S)-isomer showed the reversed Cotton effects at the corresponding wavelengths (calculated at 235 and 208 nm). For the compounds with OH-20 and H-23 in the cis orientation, the (20S,23S)-isomers showed negative Cotton effects at 230 ± 2 nm and positive effects at 204 ± 2 nm (calculated at 235 and 208 nm), while the (20R,23R)-isomer showed the reversed Cotton effects at the corresponding wavelengths (calculated at 236 and 212 nm). This also demonstrated that the C-23 configuration of the γ-lactone moiety in these compounds is determining the signs of the Cotton effects, while the configuration change at C-20 induces only a variation in the intensity of the Cotton effects. Comparison of the NMR spectra of compound 8 with those of 1 (Tables 1 and 3) indicated that the only difference between these two compounds was replacement of the oxymethine (C-23) and carbonyl (C-21) units in 1 by methylene and hydroxymethyl groups in 8, respectively. This suggested that 8 had a 3,20,21-trihydroxydammar-24-ene aglycone moiety,6a,8a,11 which was further supported by HRESIMS and confirmed by 2D NMR analysis. Particularly, HMBC correlations of H2-21/C-17, C-20, and C-22; H2-23/ C-20, C-22, C-24, and C-25; and H-1′/C-3, in combination with their chemical shifts, verified the presence of hydroxy groups at C-20 and C-21 in 8. The S configuration at C-20 was proposed from the chemical shifts of C-20 (δ 76.6) and C-17 (δ 46.3).11a,12 This was confirmed by the in situ dimolybdenum ECD method,13 which was reported for the assignment of the configurations of acyclic 1,2-diols.4,14 According to the empirical rule proposed by Snatzke,13,15 the bands around 310 nm (band IV) and 400 nm (band II) in the Mo2(AcO)4induced ECD spectrum, which have the same sign as the O−C−C−O torsion angle in the favored conformation, allow for the assignment of the absolute configuration.13b,15,16 Acid hydrolysis of 8 yielded the aglycone 8a. In the Mo2(AcO)4induced ECD spectrum of 8a (Supporting Information, Figure S68), positive Cotton effects at 309 and 386 nm supported the 1378

dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products

Article

3-O-{[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]β-D-glucopyranosyl}-21-O-β-D-glucopyranosyl-(1→6)-β-D-glucopyranoside. The other known compounds were identified by comparison of spectroscopic data with reported data as gylongiposide I, gypenosides XLVIII11a and XLIX,12a (3β,20S)-3,20,21-trihydroxydammar-24-ene 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-Dxylopyranosyl-(1→3)]-β-D-glucopyranoside,12c (3β,20S)-3,19, 20,21-tetrahydroxydammar-24-ene 3-O-{[α-L-rhamnopyranosyl(1→2)]-[β-D-xylopyranosyl-(1→3)]-α-L-arabinopyranosyl}-21-Oβ-D-glucopyranoside, (3β,20S)-3,19,20,21-tetrahydroxydammar24-ene 3-O-{[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl(1→3)]-β- D -glucopyranosyl}-21-O-β- D -glucopyranoside, (3β,20S)-3,20,21-trihydroxydammar-24-ene 3-O-{[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-glucopyranosyl}-21-O-β-D-glucopyranoside, and (3β,20S)-3,20,21-trihydroxydammar-24-ene 3-O-{[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-6-O-acetylglucopyranosyl}-21-O-β-D-glucopyranoside.11b Similar dammarane derivatives from G. pentaphyllum have been reported to possess biological activities such as protein tyrosine phosphatase 1B (PTP1B) inhibitory activity8b,10 and potential apoptotic effect17 or cytotoxic activity against various cancer cell lines.18 The isolates from M. yaoshansis were tested at 10 μM in preliminary assays to assess their PTP1B inhibitory activity,19 cytotoxicity against A2780 ovary, HCT-8 colon, Bel7402 hepatoma, BGC-823 stomach, and A549 lung cancer cell lines,20 and the TNF-α secretion inhibitory activity of mouse peritoneal macrophages.21 However, all the isolates were found to be inactive in all assays.

20S configuration. Therefore, compound 8 was determined as (3β,20S)-3,20,21-trihydroxydammar-24-ene 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-α-L-arabinopyranoside. The spectroscopic data of compound 9 indicated that it was a derivative of 8 with an additional β-glucopyranosyl unit. Comparison of the NMR spectra of 9 and 8 demonstrated that the H-21a, H-21b, and C-21 resonances in 9 were deshielded by ΔδH +0.32, +0.01 and ΔδC +9.5 ppm, respectively. This suggested that the β-glucopyranosyl unit was located at C-21 in 9, which was confirmed by HMBC correlations of H-1″″/C-21 and H2-21/C-1″″. Thus, compound 9 was determined as (3β,20S)-3,20,21-trihydroxydammar-24ene 3-O-{[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl(1→3)]-α-L-arabinopyranosyl}-21-O-β-D-glucopyranoside. Spectroscopic data analysis of compound 10 (C40H66O12) indicated that it was another derivative of 8, with one less rhamnopyranosyl unit and the methyl group replaced by a formyl group (δH 10.03 and δC 205.7). In the HMBC spectrum, correlations of H-5 and H-9/C-19 and H-19/C-1, C-9, and C-10 revealed that the formyl group was located at C-10 in 10, while HMBC correlations of H-1′/C-3 and H-1″/C-3′ proved the linkage of the sugar units. Thus, compound 10 was determined as (3β,20S)-19-oxo-3,20,21-trihydroxydammar-24ene 3-O-β-D-xylopyranosyl-(1→3)-α-L-arabinopyranoside. Comparison of the spectroscopic data of compound 11 with those of 9 indicated that the methyl (CH3-19) and αarabinopyranosyl units in 9 were replaced by a formyl group and a β-glucopyranosyl moiety in 11, respectively. This was confirmed by 2D NMR analysis of 11, as well as by acid hydrolysis of 11 followed by GC analysis of the hydrolysate using the same protocol as described above. In particular, HMBC correlations of H-5 and H-9/C-19; H-19/C-1, C-9, and C-10; H-1′/C-3; H-1″/C-2′; H-1″′/C-3′; and H-1″″/C-21 verified the location of the formyl group and the linkage of the sugar units in 11. Therefore, compound 11 was defined as (3β,20S)-19-oxo-3,20,21-trihydroxydammar-24-ene 3-O-{[α-Lrhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-glucopyranosyl}-21-O-β-D-glucopyranoside. Compound 12 had the molecular formula C47H80O16 as indicated by HRESIMS and NMR data. Comparison of the NMR data of 12 and 8 suggested that the α-arabinopyranosyl group in 8 was replaced by a β-D-6-O-acetylglucopyranosyl moiety in 12. The suggestion was confirmed by 2D NMR experiments and GC analysis of the hydrolysate of 12. Particularly, in the HMBC spectrum of 12, correlations of H-1′/C-3, H-1″/C-2′, and H-1″′/C-3′ confirmed the linkage of the glycosyl moieties, while a correlation from the H2-6′ resonance to the acetyl carbonyl carbon (δC 170.6) verified the location of the acetyl group at C-6′. Thus, compound 12 was determined as (3β,20S)-3,20,21-trihydroxydammar-24-ene 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-6-O-acetylglucopyranoside. Compound 13 had the molecular formula C59H100O26, as indicated by HRESIMS and NMR data. The NMR spectra of 13 were similar to those of 12, except for the presence of resonances attributable to two additional β-glucopyranosyl units and the absence of the acetyl resonances in 13. The NMR resonance assignments for 13 were confirmed by 2D NMR data analysis. Particularly, the HMBC correlations of H-1′/C-3, H-1″/C-2′, H-1″′/C-3′, H-1″″/C-21, and H-1″″′/C-6″″ demonstrated the linkage of the sugar units in 13. Therefore, compound 13 was determined as (3β,20S)-3,20,21-trihydroxydammar-24-ene



EXPERIMENTAL SECTION

General Experimental Procedures. Optical rotations were measured using a Rudolph Research Autopol III automatic polarimeter. UV and ECD spectra were recorded using a JASCO J-815 spectropolarimeter. IR spectra were recorded using a Nicolet 5700 FTIR microscope spectrometer (FT-IR microscope transmission). 1D and 2D NMR spectra were obtained at 400, 500, or 600 MHz for 1H and 100, 125, or 150 MHz for 13C using a Varian 400, 500, or 600 MHz NMR spectrometer in pyridine-d5, acetone-d6, or CDCl3, with solvent peaks used as references. ESIMS data were measured using a Q-Trap LC/MS/MS (Turbo Ionspray Source) spectrometer. HRESIMS data were measured using an AccuToFCS JMS-T100CS spectrometer. Column chromatography was performed with HPD-100 macroporous adsorbent resin (Cangzhou Bonchem Co., Ltd., China), silica gel (200−300 mesh, Qingdao Marine Chemical Inc., China), and Sephadex LH-20 (GE Healthcare Bio-Sciences AB, Sweden). HPLC was performed using a Waters 600 controller, a Waters 600 pump, and a Waters 2487 dual λ absorbance detector with an Alltima (250 × 10 mm i.d.) preparative C18 (5 μm) column. TLC was carried out with glass precoated silica gel GF254 plates. Spots were visualized under UV light or by spraying with 7% H2SO4 in 95% aqueous EtOH followed by heating. Plant Material. Roots of Machilus yaoshansis were collected at Dayao Mountain, Guangxi, China, in December 2007. The plant was identified by Mr. Guang-Ri Long (Guangxi Forest Administration, Guangxi 545005, China). A voucher specimen (no. 07114) was deposited at the Herbarium of Guangxi Forest Administration, China. Extraction and Isolation. The air-dried roots of M. yaoshansis (10 kg) were powdered and extracted with 95% aqueous EtOH (3 × 15 L) at room temperature (3 × 48 h). The EtOH extract was evaporated under reduced pressure to yield a dark brown residue (1050 g). The residue was suspended in H2O (5 L) and partitioned with EtOAc (5 × 5 L). The aqueous phase was loaded onto an HPD-100 macroporous adsorbent resin (1500 g, dry weight) column. Successive elution with H2O, 30% EtOH(aq), 70% EtOH(aq), and 95% EtOH(aq) 1379

dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products

Article

IR νmax 3457, 3403, 1725, 1620, 1447, 1388 cm−1; 1H NMR (pyridined5, 500 MHz) data, see Table 2; 13C NMR (pyridine-d5, 125 MHz) data, see Table 3; ESIMS m/z 965 [M + Na]+; HRESIMS m/z 965.5454 [M + Na]+ (calcd for C49H82O17Na, 965.5450). (3β,20S)-3,20,21-Trihydroxydammar-24-ene 3-O-{[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-glucopyranosyl}21-O-β-D-glucopyranosyl-(1→6)-β-D-glucopyranoside (13): amorphous powder; [α]20D −9.3 (c 0.82, MeOH); IR νmax 3390, 1683, 1646, 1451, 1374 cm−1; 1H NMR (pyridine-d5, 500 MHz) data, see Table 2; 13C NMR (pyridine-d5, 125 MHz) data, see Table 3; ESIMS m/z 1223 [M − H]−, 1247 [M + Na]+; HRESIMS m/z 1247.6408 [M + Na]+ (calcd for C59H100O26Na, 1247.6400). Acid Hydrolysis of 1−13. Compounds 1−13 (5−10 mg) were individually hydrolyzed with 1 N HCl−dioxane (1:1, 3 mL) at 60 °C for 6 h. After dilution with H2O (5 mL), the reaction mixture was extracted with EtOAc to yield separate EtOAc and H2O phases. The organic layer was concentrated, and the residue was purified by HPLC using 90% MeOH in H2O to afford 1a and 1b from 1−7, and 8a from 8, 9, 12, and 13. 1a: [α]20D −9.5 (c 0.1, MeOH); ECD (MeOH) 208 (Δε −1.06), 226 (Δε +1.82); 1H NMR (acetone-d6 and CDCl3, 600 MHz) and 13C NMR (CDCl3, 125 MHz) data, see Supporting Information, Table S1; ESIMS m/z 495 [M + Na]+. 1b: [α]20D +41.3 (c 0.3, MeOH); ECD (MeOH) 206 (Δε +0.36), 230 (Δε −2.11); 1H NMR (acetone-d6 and CDCl3, 600 MHz) and 13C NMR (CDCl3, 125 MHz) data, see Supporting Information, Table S1; ESIMS m/z 495 [M + Na]+. 8a: [α]20D +32.1 (c 0.3, MeOH); Mo2(OAc)4-induced ECD (DMSO) 272 (Δε′ −0.26), 309 (Δε′ +0.59), 359 (Δε′ +0.09), 386 (Δε′ +0.11); 1H NMR (CDCl3, 400 MHz) data, see Supporting Information, Table S2; 13C NMR (CDCl3, 100 MHz) data, see Supporting Information, Table S3; ESIMS m/z 483 [M + Na]+, 459 [M − H]−. The H2O layer was evaporated under reduced pressure. After addition of H2O (5 mL), the acidic solution was evaporated again, and this procedure was repeated until a neutral solution was obtained. The neutral solution was evaporated and dried in vacuo to furnish a monosaccharide residue. The residue was dissolved in pyridine (0.5 mL), and 2 mg of L-cysteine methyl ester hydrochloride was added. The mixture was maintained at 60 °C for 2 h, evaporated under a stream of N2, and dried in vacuo. Next, 0.2 mL of N-trimethylsilylimidazole was added, and the resultant reaction mixture was maintained at 60 °C for 1 h. The mixture was partitioned between n-hexane and H2O (2 mL each), and the n-hexane extract was analyzed by GC-MS under the following conditions: capillary column, DB-5 (30 m × 0.25 mm × 0.25 μm); detector, FID; detector temperature, 280 °C; injection temperature, 250 °C; initial temperature, 100 °C for 2 min and subsequent increase to 280 °C at a rate of 10 °C/min; final temperature, 280 °C for 5 min; carrier, N2 gas. The absolute configurations of the sugars isolated from the hydrolysates of 1−13 were determined by comparing the retention times of their trimethylsilyl-L-cysteine derivatives with those of authentic sugars prepared by a similar procedure. The retention times of the trimethylsilylL-cysteine derivatives of the sugars were as follows: D-glucose, 19.55 min; D-xylopyranose, 17.65 min; L-rhamnopyranose, 18.38 min; and Larabinopyranose, 17.79 min. ECD Calculation. Conformational analyses of 1a and the (20R,23S)-isomer were carried out via Monte Carlo searching in the MMFF94 molecular mechanics force field using the SPARTAN 08 software.22 The two lowest energy conformers for 1a and the 12 lowest energy conformers for the (20R,23S)-isomer (Supporting Information, Figure S2), whose relative energies were within 2 kcal/ mol, were considered for further DFT calculations. Subsequently, the conformers were reoptimized using DFT at the B3LYP/6-31G(d) level in the gas phase with the Gaussian 09 program.23 The B3LYP/ 6-31G(d) harmonic vibrational frequencies were further calculated to confirm their stability. The energies, oscillator strengths, and rotational strengths of the first two electronic excitations of the conformers were calculated using TDDFT methodology at the B3LYP/6-311++G(2d,2p) level in the gas phase, and the ECD spectra were simulated by the GaussSum 2.25 program (σ = 0.8 eV).24 To obtain the final spectra of 1a and the (20R,23S)-isomer, the simulated spectra of the corresponding lowest energy conformations were averaged according

(10 L each) and solvent removal yielded four corresponding fractions. The fraction (170 g) eluted by 70% EtOH(aq) was separated over silica gel, with elution using a gradient of increasing MeOH concentration in CHCl3 (2−100%) to give eight fractions (A−H). Fraction C (28 g) was fractionated via RP-MPLC using a preparative C18 (5 μm) column, with elution using a gradient of increasing MeOH concentration (0−85%) in H2O to give fractions C1−C7. Fraction C5 (2.9 g) was further chromatographed over Sephadex LH-20 (MeOH− H2O, 1:1) to afford fractions C5‑1−C5‑4. Subsequent separation of C5‑2 by RP-HPLC using MeOH−H2O (85:15) as the mobile phase afforded 1 (36 mg), 9 (15 mg), and 12 (9 mg), and separation of C5‑3 gave 2 (16 mg), 3 (33 mg), 8 (17 mg), and 10 (15 mg). Purification of C5‑4 by RP HPLC using MeOH−H2O (83:17) as the mobile phase gave 11 (12 mg) and 13 (35 mg). (3β,20S,23R)-3,20,23-Trihydroxydammar-24-en-21-oic acid21,23-lactone 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-α-L-arabinopyranoside (1): amorphous powder; [α]20D −14.8 (c 0.49, MeOH); ECD (MeOH) 204 (Δε −3.46), 227 (Δε +1.66) nm; IR νmax 3383, 1760, 1647, 1549, 1449, 1377 cm−1; 1H NMR (pyridine-d5, 400 MHz) data, see Table 1; 13C NMR (pyridined5, 100 MHz) data, see Table 3; ESIMS m/z 905 [M + Na]+; HRESIMS m/z 905.4832 [M + Na]+ (calcd for C46H74O16Na, 905.4874). (3β,20S,23S)-3,20,23-Trihydroxydammar-24-en-21-oic acid21,23-lactone 3-O-[α- L -rhamnopyranos yl-(1→2)]-[β- D xylopyranosyl(1→3)]-α-L-arabinopyranoside (2): amorphous powder; [α]20D +7.1 (c 0.10, MeOH); ECD (MeOH) 203 (Δε +1.76), 231 (Δε −1.62) nm; IR νmax 3412, 1753, 1666, 1647, 1547, 1448, 1378 cm−1; 1H NMR (pyridine-d5, 400 MHz) data, see Table 1; 13C NMR (pyridine-d5, 100 MHz) data, see Table 3; ESIMS m/z 905 [M + Na]+; HRESIMS m/z 905.4914 [M + Na]+ (calcd for C46H74O16Na, 905.4874). (3β,20S,23S)-3,20,23-Trihydroxydammar-24-en-21-oic acid21,23-lactone 3-O-β-D-xylopyranosyl-(1→3)-β-D-glucopyranoside (3): amorphous powder; [α]20D +13.8 (c 0.65, MeOH); ECD (MeOH) 205 (Δε +0.62), 231 (Δε −1.31) nm; IR νmax 3406, 1756, 1642, 1450, 1377 cm−1; 1H NMR (pyridine-d5, 500 MHz) data, see Table 1; 13C NMR (pyridine-d5, 125 MHz) data, see Table 3; ESIMS m/z 789 [M + Na]+; HRESIMS m/z 789.4394 [M + Na]+ (calcd for C41H66O13Na, 789.4401). (3β,20S)-3,20,21-Trihydroxydammar-24-ene 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-α-L-arabinopyranoside (8): amorphous powder; [α]20D +6.0 (c 0.42, MeOH); IR νmax 3401, 1643, 1562, 1451, 1377 cm−1; 1H NMR (pyridine-d5, 500 MHz) data, see Table 1; 13C NMR (pyridine-d5, 125 MHz) data, see Table 3; ESIMS m/z 893 [M + Na]+; HRESIMS m/z 893.5279 [M + Na]+ (calcd for C46H78O15Na, 893.5238). (3β,20S)-3,20,21-Trihydroxydammar-24-ene 3-O-{[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-α-L-arabinopyranosyl}-21-O-β-D-glucopyranoside (9): amorphous powder; [α]20D −13.6 (c 0.80, MeOH) IR νmax 3420, 1642, 1562, 1451, 1375 cm−1; 1H NMR (pyridine-d5, 500 MHz) data, see Table 1; 13C NMR (pyridine-d5, 125 MHz) data, see Table 3; ESIMS m/z 1055 [M + Na]+; HRESIMS m/z 1055.5735 [M + Na]+ (calcd for C52H88O20Na, 1055.5767). (3β,20S)-19-Oxo-3,20,21-trihydroxydammar-24-ene 3-O-β-D-xylopyranosyl-(1→3)-α-L-arabinopyranoside (10): amorphous powder; [α]20D +20.9 (c 0.85, MeOH); IR νmax 3379, 1700, 1646, 1594, 1443, 1377 cm−1; 1H NMR (pyridine-d5, 400 MHz) data, see Table 2; 13C NMR (pyridine-d5, 100 MHz) data, see Table 3; ESIMS m/z 761 [M + Na]+; HRESIMS m/z 761.4424 [M + Na]+ (calcd for C40H66O12Na, 761.4452). (3β,20S)-19-Oxo-3,20,21-trihydroxydammar-24-ene 3-O-{[α-Lrhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-glucopyranosyl}-21-O-β-D-glucopyranoside (11): amorphous powder; [α]20D +2.2 (c 0.38, MeOH); IR νmax 3406, 1702, 1647, 1448, 1378 cm−1; 1H NMR (pyridine-d5, 500 MHz) data, see Table 2; 13C NMR (pyridine-d5, 125 MHz) data, see Table 3; ESIMS m/z 1099 [M + Na]+; HRESIMS m/z 1099.5642 [M + Na]+ (calcd for C53H88O22Na,1099.5665). (3β,20S)-3,20,21-Trihydroxydammar-24-ene 3-O-[α-L-rhamnopyranosyl-(1→2)]-[β-D-xylopyranosyl-(1→3)]-β-D-6-O-acetylglucopyranoside (12): amorphous powder; [α]20D −5.3 (c 0.32, MeOH); 1380

dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products

Article

Zhang, Y.; Zi, J.; Song, W.; Fan, X.; Liu, Y.; Yang, Y.; Shi, J. Org. Lett. 2011, 13, 2856−2859. (c) Liu, M.; Lin, S.; Gan, M.; Chen, M.; Li, L.; Wang, S.; Zi, J.; Fan, X.; Liu, Y.; Si, Y.; Yang, Y.; Chen, X.; Shi, J. Org. Lett. 2012, 14, 1004−1007. (4) Gan, M.; Liu, M.; Liu, B.; Lin, S.; Zhang, Y.; Zi, J.; Song, W.; Ye, F.; Chen, X.; Shi, J. J. Nat. Prod. 2011, 74, 2431−2437. (5) (a) Hara, S.; Okabe, H.; Mihashi, K. Chem. Pharm. Bull. 1987, 35, 501−506. (b) Kinjo, J.; Araki, K.; Fukui, K.; Higuchi, H.; Ikeda, T.; Nohara, T.; Ida, Y.; Takemoto, N.; Miyakoshi, M.; Shoji, J. Chem. Pharm. Bull. 1992, 42, 3269−3273. (6) (a) Piacente, S.; Pizza, C.; De Tommasi, N.; De Simone, F. J. Nat. Prod. 1995, 58, 512−519. (b) Yin, F.; Hu, L.-H. Helv. Chim. Acta 2005, 88, 1126−1134. (c) Diome, C.; Mitaine-Offer, A.-C.; Miyamoto, T.; Delaude, C.; Mirjolet, J.-F.; Duchamp, O.; Lacaille-Dubois, M.-A. J. Nat. Prod. 2007, 70, 1680−1682. (7) (a) Ding, Y.; Li, X.-C.; Ferreira, D. J. Org. Chem. 2007, 72, 9010− 9017. (b) Gan, L.-S.; Zheng, Y.-L.; Mo, J.-X.; Liu, X.; Li, X.-H.; Zhou, C.-X. J. Nat. Prod. 2009, 72, 1497−1501. (c) Li, X. C.; Ferreira, D.; Ding, Y. Curr. Org. Chem. 2010, 14, 1678−1697. (d) Gan, M.; Zheng, X.; Gan, L.; Guan, Y.; Hao, X. Q.; Liu, Y.; Si, S.; Zhang, Y.; Yu, L.; Xiao, C. J. Nat. Prod. 2011, 74, 1142−1147. (e) Neff, S. A.; Lee, S. U.; Asami, Y.; Ahn, J. S.; Oh, H.; Baltrusaitis, J.; Gloer, J. B.; Wicklow, D. T. J. Nat. Prod. 2012, 75, 464−472. (8) (a) Shi, L.; Cao, J.-Q.; Li, W.; Zhao, Y.-Q. Helv. Chim. Acta 2010, 93, 1785−1794. (b) Xu, J.-Q.; Shen, Q.; Li, J.; Hu, L.-H. Biorg. Med. Chem. 2010, 18, 3934−3939. (9) (a) Asakawa, J.; Kasai, R.; Yamasaki, K.; Tanaka, O. Tetrahedron 1977, 33, 1935−1939. (b) Junxian, W.; Liangyu, C.; Jufen, W.; Friedrichs, E.; Jores, M.; Puff, H.; Chen Wei, s.; Breitmaier, E. Planta Med. 1982, 45, 167−171. (c) Tori, M.; Matsuda, R.; Sono, M.; Asakawa, Y. Magn. Reson. Chem. 1988, 26, 581−590. (10) Hung, T. M.; Hoang, D. M.; Kim, J. C.; Jang, H.-S.; Ahn, J. S.; Min, B.-S. J. Ethnopharmacol. 2009, 124, 240−245. (11) (a) Yin, F.; Hu, L.; Pan, R. Chem. Pharm. Bull. 2004, 52, 1440− 1444. (b) Yin, F.; Hu, L.; Lou, F.; Pan, R. J. Nat. Prod. 2004, 67, 942− 952. (12) (a) Takemoto, T.; Arihara, S.; Yoshikawa, K.; Hino, K.; Nakajima, T.; Okuhira, M. Yakugaku Zasshi 1984, 104, 1155−1162. (b) Kizu, H.; Koshijima, M.; Tomimori, T. Chem. Pharm. Bull. 1985, 33, 3176−3181. (c) Shi, L.; Cao, J. Q.; Li, W.; Zhao, H.; Zhao, Y. Q. Chin. Chem. Lett. 2010, 21, 699−701. (13) (a) Frelek, J.; Geiger, M.; Voelter., W. Curr. Org. Chem. 1999, 3, 117−146. (b) Di Bari, L.; Pescitelli, G.; Pratelli, C.; Pini, D.; Salvadori, P. J. Org. Chem. 2001, 66, 4819−4825. (c) Frelek, F.; Klimek, A.; Ruskowska, P. Curr. Org. Chem. 2003, 7, 1081−1104. (14) (a) Wu, X.; Lin, S.; Zhu, C.; Yue, Z.; Yu, Y.; Zhao, F.; Liu, B.; Dai, J.; Shi, J. J. Nat. Prod. 2010, 73, 1294−1300. (b) Liu, J.; Liu, Y.; Si, Y.; Yu, S.; Qu, J.; Xu, S.; Hu, Y.; Ma, S. Steroids 2009, 74, 51−61. (c) Dong, S.-H.; Zhang, C.-R.; Dong, L.; Wu, Y.; Yue, J.-M. J. Nat. Prod. 2011, 74, 1042−1048. (15) Frelek, J.; Snatzke, G. Fresenius J. Anal. Chem. 1983, 316, 261− 264. (16) (a) Frelek, J.; Ruskowska, P.; Suszczynska, A.; Szewczyk, K.; Osuch, A.; Jarosz, S.; Jagodzinski, J. Tetrahedron: Asymmetry 2008, 19, 1709−1713. (b) Górecki, M.; Jabłońska, E.; Kruszewska, A.; Suszczyńska, A.; Urbańczyk-Lipkowska, Z.; Gerards, M.; Morzycki, J. W.; Szczepek, W. J.; Frelek, J. J. Org. Chem. 2007, 72, 2906−2916. (17) (a) Chen, J.-C.; Lu, K.-W.; Lee, J.-H.; Yeh, C.-C.; Chung, J.-G. Anticancer Res. 2006, 26, 4313−4326. (b) Wang, Q.-F.; Chiang, C.-W.; Wu, C.-C.; Cheng, C.-C.; Hsieh, S.-J.; Chen, J.-C.; Hsieh, Y.-C.; Hsu, S.-L. Planta Med. 2007, 73, 535−544. (18) Ky, P. T.; Huong, P. T.; My, T. K.; Anh, P. T.; Kiem, P. V.; Van Minh, C.; Cuong, N. X.; Thao, N. P.; Nhiem, N. X.; Hyun, J.-H.; Kang, H.-K.; Kim, Y. H. Phytochemistry 2010, 71, 994−1001. (19) Wang, Y.; Shang, X. Y.; Wang, S. J.; Mo, S. Y.; Li, S.; Yang, Y. C.; Ye, F.; Shi, J. G.; He, L. J. Nat. Prod. 2007, 70, 296−299. (20) Mo, S.; Wang, S.; Zhou, G.; Yang, Y.; Li, Y.; Chen, X.; Shi, J. J. Nat. Prod. 2004, 67, 823−828.

to the Boltzmann distribution theory, in which the relevant Gibbs free energies (G) were adopted. Conformational analyses in the MMFF94 force field showed three and eight lowest energy conformers for 1b and the (20R,23R)-isomer, respectively, whose relative energies were within 2.0 kcal/mol (Supporting Information, Figure S4). The conformers were reoptimized using DFT at the B3LYP/6-31G(d) level in the gas phase. The energies, oscillator strengths, and rotational strengths of the first four electronic excitations for 1b and the first two electronic excitations for the (20R,23R)-isomer were calculated using TDDFT methodology at the B3LYP/6-311++G(2d,2p) level in the gas phase, and the ECD spectra were simulated by the GaussSum 2.25 software (σ = 0.6 eV). The final spectra of 1b and the (20R,23R)-isomer were obtained by averaging the corresponding spectra according to their relative conformational Gibbs free energies PTP1B Inhibition Assay. See ref 19. Cells, Culture Conditions, and Cell Proliferation Assay. See ref 20. TNF-α Secretion Inhibition Assay. See ref 21.



ASSOCIATED CONTENT

* Supporting Information S

NMR data for compounds 1a and 1b (Table S1) and for 4−7 and 8a (Tables S2 and S3). ECD spectra calculation details of 1a and the (20R,23S)-isomer and 1b and the (20R,23R)isomer. Copies of MS, ECD, IR, and NMR spectra of compounds 1−13. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*Tel: 86-10-83154789. Fax: 86-10-63017757. E-mail: lsznn@ imm.ac.cn, [email protected]. Author Contributions §

These authors contributed equally to this study.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We thank Professor J. Zhang (Institute of Medical Science, School of Medicine, Shanghai Jiaotong University) for his kind help on the calculation of ECD spectra. Financial support from the National Natural Sciences Foundation of China (NNSFC; grant nos. 30825044, 81102341, 20932007, and 21132009), the Program for Changjiang Scholars and Innovative Research Team in University (PCSIRT, grant no. IRT1007), and the National Science and Technology Project of China (no. 2011ZX09307-002-01) is acknowledged.



REFERENCES

(1) (a) Jiangsu New Medical College. Dictionary of Traditional Chinese Medicine; Shanghai Science and Technology Publishing House: Shanghai, 1977; pp 114, 1009, and 1423. (b) Cheng, M.-J.; Tsai, I.-L.; Lee, S.-J.; Jayaprakasam, B.; Chen, I.-S. Phytochemistry 2005, 66, 1180−1185. (c) Park, E. Y.; Shin, S. M.; Ma, C. J.; Kim, Y. C.; Kim, S. G. Planta Med. 2005, 71, 393−398. (d) Giang, P. M.; Son, P. T.; Matsunami, K.; Otsuka, H. Chem. Pharm. Bull. 2006, 54, 380−383. (2) (a) Cheng, W.; Zhu, C.; Xu, W.; Fan, X.; Yang, Y.; Li, Y.; Chen, X.; Wang, W.; Shi, J. J. Nat. Prod. 2009, 72, 2145−2152. (b) Li, Y.; Cheng, W.; Zhu, C.; Yao, C.; Xiong, L.; Tian, Y.; Wang, S.; Lin, S.; Hu, J.; Yang, Y.; Guo, Y.; Yang, Y.; Li, Y.; Yuan, Y.; Chen, N.; Shi, J. J. Nat. Prod. 2011, 74, 1444−1452. (3) (a) Liu, M.; Lin, S.; Wang, Y.; He, W. e.; Li, S.; Wang, S.; Yang, Y.; Shi, J. Org. Lett. 2007, 9, 129−132. (b) Liu, M.; Gan, M.; Lin, S.; 1381

dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382

Journal of Natural Products

Article

(21) Lin, S.; Wang, S.; Liu, M.; Gan, M.; Li, S.; Yang, Y.; Wang, Y.; He, W.; Shi, J. J. Nat. Prod. 2007, 70, 817−823. (22) Spartan 08; Wavefunction, Inc.: Irvine, CA. (23) Gaussian 09, Revision A.1; Gaussian, Inc.: Wallingford, CT, 2009. A full list of authors can be found in the Supporting Information. (24) O’Boyle, N. M.; Tenderholt, A. L.; Langner, K. M. J. Comput. Chem. 2008, 29, 839−845.

1382

dx.doi.org/10.1021/np300310a | J. Nat. Prod. 2012, 75, 1373−1382